A key factor in deciding what satellite or sensor to use for your data collection is going to be the temporal resolution, which is essentially just how often am I able to obtain an image of a particular area from that sensor? So the temporal resolution is the time interval between measurements, how often a sensor is able to revisit an area. Now, this depends on the sensor platform. So if you have your own plane, and you have a sensor on that plane, and you have unlimited money and staff, then you can go fly every day if you want, and take images every hour of every day. So that would be a very high temporal resolution, because you're able to revisit that area very frequently. However, if you have, say, a satellite that's orbiting the earth, that satellite may only come back to that exact same spot on the ground, say, every few days. So that would be a much lower temporal resolution. So when you're looking for data, that may be a factor in terms of the types of images you're looking for or the types of things you're trying to map. So if you want to be able to view the same area repeatedly to detect changes, or as we would refer to it as change detection, things like the spread of a forest fire, or the extent of flooding. Those are things that may be happening over a matter of hours or days, so you'd want something with a fairly high temporal resolution. But if you're looking at something that's happening over a much longer period of time. Say urban sprawl, or urban growth, then you might have something with a much lower temporal resolution that's perfectly acceptable, where it may only be needed that you're collecting imagery, say, once a year. And clouds can affect the results as well. So even if you have a fairly high temporal resolution, say, something that is able to revisit an area every 16 days, that's what it would be for, say, a Landsat satellite. If that area is often covered in clouds then it's going to be more difficult to be able to get a useful image out of that. So cloud cover in itself does not directly relate to temporal resolution that's separate but it will be a factor in that if you need to be able to get cloud-free imagery. Then you want to be able to have a fairly high temporal resolution, hopefully, so that you have more chances to be able to find a cloud-free day. So here we have an image that's from a nice website. I've got the link down here that you can go to if you are interested in looking at it, is that what it's mapping here is the locations of all of the objects that are orbiting the earth. It's incredible, there's thousands of them. But the main patterns that I want you to see, for the purposes of what we're talking about, is the fact that we have a lot of satellites that are in what's known as a near-polar orbit. So they are very close to the surface of the Earth. And then we have a ring of satellites that are much farther away and they all happen to be around the equator. And those are referred to as a Geostationary orbit. So the ones that are close to the earth, again, there are other types of orbits. You can see there's all kinds of other little specks out there, we're not going to pay attention to those for now. What we're mostly interested in is the fact that we have sun satellites in the near polar orbit and other ones that are in the geostationary orbit. So as the earth is rotating, this is my little PowerPoint animation, is that the ones that are close to the earth are going in more or less a north-south direction around the poles. And then the geostationary orbits are much larger since they're going over a much larger distance. So this is the actual website for the image that I was just showing you. And this is interactive, so we can actually rotate the Earth here, we can zoom in and out. And so I'm hoping what you can see here is, as I rotated it, is that there's this ring of satellites here. And these are the ones in a geostationary orbit around the earth, and then we have the ones closer in the near polar orbit. Zoom in a bit here, and there are presets, so we can look at things like the space station or in this case the Landsat satellite. So that's one that's very popular for doing land cover mapping and let's see if we can find it. So here we have Landsat 7, you can see it's over Central America right now, and we have the path of the orbit. So this is in a near polar orbit, it's not exactly going over the poles, but it's still referred to as a near polar orbit. So you can see that the satellite is at an orbit of 703 kilometers above the Earth, so that's good to know. So the satellites that we use for remote sensing and mapping are in a sun-synchronous orbit. That means that they're passing over the equator at the same time each time they do the pass, so usually they're often around 10:30 in the morning. So that wherever it's taking an image, not just at the equator, but that's the time that it's specified when it's at that particular position. But as it's moving over the surface of the Earth, the angle of the orbit is such that it's always at the same time of day when it's taking an image. And that makes it much easier to be able to compare imagery from different days, and even across a season. So that when we see differences between these images, we know that it's not because of the angle of the sun or things like that. It's that we're taking that sort of variable out, as we're able to look at them as though it's the same time of day for each image. So these satellites are in an orbit that's around 700 km. They move north to south as the Earth is rotating counterclockwise or around the pole, sorry. And so on each pass there's a different part of the Earth in that sense. And I'm hoping that this diagram kind of helps with that. So the idea here, as I'm sure you can probably see, is that this red line is tracing the orbit of the satellite as it's going around the Earth in this kind of south to north direction. But the Earth is rotating as it's going underneath the satellite. And so when the satellite comes back around to the Equator again, or wherever it happens to be, so let's say it's starting here. When it comes back around, it's over here, because the Earth has rotated while the satellite is traveling. And so then the second orbit it goes around, it's in another place again, and in the third orbit, it's in another place again. And so you have this kind of effect that the satellite is traveling this way as the Earth is rotating this way, okay? You get the idea? All right, so I think this is something that when you think about it, it's like, yeah, that makes sense. But it may not occur to you at first is that the satellite, with this type of orbit, is not coming back around to exactly the same position. The orbit is not rotating with the earth. The orbit is stationary, it's staying in the same position relative to the sun. Which is why we have the same relative time of date that the images are being taken, but the earth is rotating underneath. And the reason I mention this is that's why, for a satellite, say like Landsat 8, it takes 16 days for it to be able to actually come back to these same location. So in other words you have to have that coincidence between where the satellite is and where the Earth's rotates. And that's how many days it takes for it to come back to that exact same location to be able to revisit that same cell, that same part of the Earth. So this is my little diagram here to show how this works, is that we have the satellite moving south to north this way, and then the Earth is rotating underneath. And so we're getting that, I just hope this kind of helps you visualize what's happening, is that the Earth is rotating, the satellite is moving. And it's collecting information along that path, or collecting data, sorry, along that path as it goes. This is a great website. I love this because you can actually see satellites collecting data, if you're lucky, in real-time. But even if it's not real-time, well, actually, this one here looks like this is being recorded now from Landsat 8. So this to me is a fantastic way of getting a sense of how fast the satellite's moving, and how much data is being collected as it's going along. So this is from the United States Geological Survey, or the USGS, and this is just a wonderful way of being able to visualize. So for example, you can see here that wherever it is now it's going over an area that's covered in clouds. So it's a really good way of realizing that if there's clouds there we can't really see much unless you're mapping the clouds themselves, in which case this is awesome. But if you're not, then this isn't going to be the most useful. But it does give you a sense of how fast that satellite is actually moving over the earth and how much data is being collected. So here we can see Landsat 7 collecting data over, looks like British Columbia. These are the Rocky Mountains. And yep, there's Nelson. So this is a great way of being able to visualize how fast that satellite is moving over the earth and how much data is being collected at a given time. So that satellite is actually breaking up what it sees into individual cells, sensing the amount of light that's being reflected from each of those, and then recording that. So that data's then transmitted back to the Earth. It's assembled into stripes, let's say, which then turn into a grid, which then we can assign colors to and we can able to visually interpret as an actual image. So we can see things like Spokane, so rivers, mountains, areas with clouds, built up areas. So I love this, it's kind of hypnotic, I could spend all day just sitting here staring at it. But it's a wonderful way of being able to really see for yourself how the data are really collected. So a common application related to temporal resolution is change detection. So I'm just going to show you some examples of change detection over time to kind of give you a sense of what this can look like. So for example, here we have the area around Fort McMurray in Alberta, Canada, and this image was taken on April 17th, 2016 And this image was taken about two weeks later on May 3rd, 2016. And so, you can see there’s a large area of wildfire damage that occurred during that time. So, that was a massive wildfire, it destroyed over 1,600 structures, it forced the largest evacuation on record in Canada involving more than 80,000 people. So it was a huge ordeal for these people, and it's interesting that we can map the exact location of the damage of this using the satellite imagery. Here is another example of a wild fire. This is Anderson Creek fire. So this is before, this is on March 14, 2016. And this April 7th, 2016, so about three weeks later. And you can see the massive size of the area that was damaged during that time. So that was over 57,000 acres that were burnt during that period as part of that fire event. So very dramatic, and again, it's very useful to be able to map the location of that, get a sense of the size of the damage, the areas. And if you have before imagery and after, you can also look at things like what was there before it was burned? Was it a habitat for particular types of animals? Was it a built up area? What needs to be rebuilt? What can be left to rebuild on its own ecologically? So you get the idea there. Here we have southern Australia, this was October, 2010. And this was how the area was flooded. This is two months later, so over that period there was a massive flood that was the largest in decades. It affected over 200,000 people, cut off 22 towns. 75% of the area's coal mines were shut down, and it had a devastating effect on the country's wheat crop. And there was an enormous amount of damage that was done. So you get a sense here in terms of temporal resolution, it's not to say that the satellite only returned two months later. But that was when they were able to get a clean image, probably one that was cloud-free, and be able to see the extent of the damage. This is an interesting one here. This is Mount Pinatubo in the Philippines. And so this is an image from 1992, so this was shortly after the volcano had erupted. And you can see the area that has been completely wiped clean of vegetation at that time. Here's the same area 25 years later, and you can see how much the vegetation has regrown in that area. And how ecologically it's been able to recover from that massive volcanic event. I like this one, too. This is in Saudi Arabia, and this image is from 1986, and this is 30 years later. And look at the effect of being able to irrigate those areas in terms of being able to develop agriculture in that area. I haven't been to Saudi Arabia, but I think it's kind of fascinating region, and it's just fascinating to me that they are, in such a dry air desert climate, be able to grow crops there. Just as an example of revisit time, two of the most popular satellites that are currently in use are Landsat 7 and Landsat 8. They both have a revisit time of 16 days each. And when Landsat 8 was launched, they made sure to put it into an orbit that was opposite to Landsat 7, so that between the two of them, we're able to sense the same area every eight days. So even though the revisit time of each of them individually is 16 days, between the two of them, we can revisit an area every 8 days, which is pretty good. So far, we've been talking about satellites that are in the near-polar orbit, that are close to the earth at an altitude of about 700 kilometers. Now, let's look at ones that are much, much farther away. So these are geostationary satellites. They are at an altitude of about 36,000 kilometers. And the whole trick here that they use, the people that put these up there, is that they have the satellite orbiting the earth at the same speed that the earth is rotating. And so the effect is that the satellite appears as though it's over the same part of the earth all the time. So this is really useful for things like weather satellites or communications. To be able to map things, or to have that satellite over the same location all the time or continuously. So here's my little PowerPoint animation that shows the Earth rotating and the satellite orbiting the Earth at the same speed of the rotation. So it appears as though the satellite is always over the same part of the world. So in this case, it would be North America. This is an amazing example of what can be done at a high temporal resolution from a weather satellite. So this is a satellite that's always over the same location. So in other words, the temporal resolution really is only limited by how fast that sensor can keep taking images. It doesn't have to go around the Earth and come back to the same area, it's always over the same location, so it can take images as fast as technically possible. In this case, this is a sensor that senses lightning, and it actually can sense it 500 times per second. So this is a video that was put together by speeding up those frames. That shows lightning strikes or lightning events over the Amazon River basin over a period of time. So you can see them lighting up there, and as the sun goes down or goes away, they become more visible. And so this is a way of being able to demonstrate an extremely high temporal resolution by having that advantage of being able to have a satellite in geostationary orbit. And because it's much farther away as well, you're able to sense a much larger area than you would with a near-polar satellite, which is more closer to the earth and has to sense a much smaller part of the earth at one time.
